transfer-meta-low-shot-continual-or-long-tail-learning.md

CVPR-2023-Papers

Application
New collections

Transfer, Meta, Low-Shot, Continual, or Long-Tail Learning

Title	Repo	Paper	Video
Dynamically Instance-Guided Adaptation: A Backward-Free Approach for Test-Time Domain Adaptive Semantic Segmentation
DETR With Additional Global Aggregation for Cross-Domain Weakly Supervised Object Detection	➖
Mind the Label Shift of Augmentation-Based Graph OOD Generalization			➖
Long-Tailed Visual Recognition via Self-Heterogeneous Integration With Knowledge Excavation
Understanding and Improving Visual Prompting: A Label-Mapping Perspective
A Whac-a-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Others
Improved Distribution Matching for Dataset Condensation			➖
Divide and Adapt: Active Domain Adaptation via Customized Learning
Class Relationship Embedded Learning for Source-Free Unsupervised Domain Adaptation
Diversity-Aware Meta Visual Prompting
Uncertainty-Aware Optimal Transport for Semantically Coherent Out-of-Distribution Detection
Zero-Shot Object Counting			➖
Learning With Fantasy: Semantic-Aware Virtual Contrastive Constraint for Few-Shot Class-Incremental Learning
Distribution Shift Inversion for Out-of-Distribution Prediction
Endpoints Weight Fusion for Class Incremental Semantic Segmentation	➖
Promoting Semantic Connectivity: Dual Nearest Neighbors Contrastive Learning for Unsupervised Domain Generalization	➖		➖
Class-Conditional Sharpness-Aware Minimization for Deep Long-Tailed Recognition
Meta-Causal Learning for Single Domain Generalization
VoP: Text-Video Co-Operative Prompt Tuning for Cross-Modal Retrieval
Learning Imbalanced Data With Vision Transformers			➖
Sharpness-Aware Gradient Matching for Domain Generalization			➖
Geometry and Uncertainty-Aware 3D Point Cloud Class-Incremental Semantic Segmentation			➖
Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentation
Regularizing Second-Order Influences for Continual Learning
I2MVFormer: Large Language Model Generated Multi-View Document Supervision for Zero-Shot Image Classification			➖
FREDOM: Fairness Domain Adaptation Approach to Semantic Scene Understanding
Dense Network Expansion for Class Incremental Learning
Batch Model Consolidation: A Multi-Task Model Consolidation Framework
DiGeo: Discriminative Geometry-Aware Learning for Generalized Few-Shot Object Detection
ALOFT: A Lightweight MLP-Like Architecture With Dynamic Low-Frequency Transform for Domain Generalization
ZegCLIP: Towards Adapting CLIP for Zero-Shot Semantic Segmentation
DiGA: Distil to Generalize and then Adapt for Domain Adaptive Semantic Segmentation
Adjustment and Alignment for Unbiased Open Set Domain Adaptation
Adapting Shortcut With Normalizing Flow: An Efficient Tuning Framework for Visual Recognition			➖
CODA-Prompt: COntinual Decomposed Attention-Based Prompting for Rehearsal-Free Continual Learning
ConStruct-VL: Data-Free Continual Structured VL Concepts Learning
Generalizing Dataset Distillation via Deep Generative Prior
Few-Shot Learning With Visual Distribution Calibration and Cross-Modal Distribution Alignment			➖
Multi-Centroid Task Descriptor for Dynamic Class Incremental Inference			➖
DAA: A Delta Age AdaIN Operation for Age Estimation via Binary Code Transformer
Bilateral Memory Consolidation for Continual Learning	➖		➖
Texts as Images in Prompt Tuning for Multi-Label Image Recognition
Learning Transformations To Reduce the Geometric Shift in Object Detection	➖
CLIP the Gap: A Single Domain Generalization Approach for Object Detection
Transfer Knowledge From Head to Tail: Uncertainty Calibration Under Long-Tailed Distribution			➖
Bi-Directional Distribution Alignment for Transductive Zero-Shot Learning
DARE-GRAM: Unsupervised Domain Adaptation Regression by Aligning Inverse Gram Matrices
LASP: Text-to-Text Optimization for Language-Aware Soft Prompting of Vision & Language Models
Open-Set Likelihood Maximization for Few-Shot Learning
WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation
Federated Domain Generalization With Generalization Adjustment
ProtoCon: Pseudo-Label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-Supervised Learning	➖		➖
DA-DETR: Domain Adaptive Detection Transformer With Information Fusion	➖
Harmonious Teacher for Cross-Domain Object Detection
AutoLabel: CLIP-Based Framework for Open-Set Video Domain Adaptation
Task Difficulty Aware Parameter Allocation & Regularization for Lifelong Learning
Revisiting Prototypical Network for Cross Domain Few-Shot Learning			➖
Federated Incremental Semantic Segmentation
Semantic Prompt for Few-Shot Image Recognition
Rethinking Gradient Projection Continual Learning: Stability / Plasticity Feature Space Decoupling	➖		➖
No One Left Behind: Improving the Worst Categories in Long-Tailed Learning	➖		➖
Meta Omnium: A Benchmark for General-Purpose Learning-To-Learn			➖
Transductive Few-Shot Learning With Prototype-Based Label Propagation by Iterative Graph Refinement
COT: Unsupervised Domain Adaptation With Clustering and Optimal Transport	➖		➖
Semi-Supervised Domain Adaptation With Source Label Adaptation
MetaMix: Towards Corruption-Robust Continual Learning With Temporally Self-Adaptive Data Transformation	➖		➖
Visual-Language Prompt Tuning With Knowledge-Guided Context Optimization			➖
Modeling Inter-Class and Intra-Class Constraints in Novel Class Discovery
Real-Time Evaluation in Online Continual Learning: A New Hope
Partial Network Cloning
Rebalancing Batch Normalization for Exemplar-Based Class-Incremental Learning	➖		➖
EcoTTA: Memory-Efficient Continual Test-Time Adaptation via Self-Distilled Regularization
Feature Alignment and Uniformity for Test Time Adaptation
Bootstrap Your Own Prior: Towards Distribution-Agnostic Novel Class Discovery
Towards Realistic Long-Tailed Semi-Supervised Learning: Consistency Is All You Need
Balanced Product of Calibrated Experts for Long-Tailed Recognition
Unsupervised Continual Semantic Adaptation Through Neural Rendering
Computationally Budgeted Continual Learning: What Does Matter?
Ground-Truth Free Meta-Learning for Deep Compressive Sampling	➖
Multi-Level Logit Distillation			➖
StyleAdv: Meta Style Adversarial Training for Cross-Domain Few-Shot Learning
MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation
On the Stability-Plasticity Dilemma of Class-Incremental Learning	➖		➖
TeSLA: Test-Time Self-Learning With Automatic Adversarial Augmentation
MHPL: Minimum Happy Points Learning for Active Source Free Domain Adaptation	➖
CIGAR: Cross-Modality Graph Reasoning for Domain Adaptive Object Detection	➖
Adaptive Plasticity Improvement for Continual Learning			➖
Achieving a Better Stability-Plasticity Trade-Off via Auxiliary Networks in Continual Learning			➖
Few-Shot Geometry-Aware Keypoint Localization
Spatio-Temporal Pixel-Level Contrastive Learning-Based Source-Free Domain Adaptation for Video Semantic Segmentation
Both Style and Distortion Matter: Dual-Path Unsupervised Domain Adaptation for Panoramic Semantic Segmentation
Bi-Level Meta-Learning for Few-Shot Domain Generalization	➖
Few-Shot Referring Relationships in Videos			➖
Exploring Data Geometry for Continual Learning	➖
Masked Images Are Counterfactual Samples for Robust Fine-Tuning
DKT: Diverse Knowledge Transfer Transformer for Class Incremental Learning
CoMFormer: Continual Learning in Semantic and Panoptic Segmentation
Global and Local Mixture Consistency Cumulative Learning for Long-Tailed Visual Recognitions
Class Attention Transfer Based Knowledge Distillation
Hard Sample Matters a Lot in Zero-Shot Quantization
Back to the Source: Diffusion-Driven Adaptation To Test-Time Corruption
SuperDisco: Super-Class Discovery Improves Visual Recognition for the Long-Tail	➖
Architecture, Dataset and Model-Scale Agnostic Data-Free Meta-Learning
Preserving Linear Separability in Continual Learning by Backward Feature Projection
Upcycling Models Under Domain and Category Shift
Class-Incremental Exemplar Compression for Class-Incremental Learning
Learning Conditional Attributes for Compositional Zero-Shot Learning			➖
BlackVIP: Black-Box Visual Prompting for Robust Transfer Learning
NoisyTwins: Class-Consistent and Diverse Image Generation Through StyleGANs			➖
Semi-Supervised Learning Made Simple With Self-Supervised Clustering			➖
Guiding Pseudo-Labels With Uncertainty Estimation for Source-Free Unsupervised Domain Adaptation
PCR: Proxy-Based Contrastive Replay for Online Class-Incremental Continual Learning
Modality-Agnostic Debiasing for Single Domain Generalization	➖
Robust Mean Teacher for Continual and Gradual Test-Time Adaptation			➖
Foundation Model Drives Weakly Incremental Learning for Semantic Segmentation	➖		➖
Hubs and Hyperspheres: Reducing Hubness and Improving Transductive Few-Shot Learning With Hyperspherical Embeddings
Robust Test-Time Adaptation in Dynamic Scenarios
Source-Free Video Domain Adaptation With Spatial-Temporal-Historical Consistency Learning	➖
Heterogeneous Continual Learning
Continual Detection Transformer for Incremental Object Detection
NIFF: Alleviating Forgetting in Generalized Few-Shot Object Detection via Neural Instance Feature Forging	➖
ViewNet: A Novel Projection-Based Backbone With View Pooling for Few-Shot Point Cloud Classification			➖
C-SFDA: A Curriculum Learning Aided Self-Training Framework for Efficient Source Free Domain Adaptation			➖
Train/Test-Time Adaptation With Retrieval	➖		➖
Dealing With Cross-Task Class Discrimination in Online Continual Learning
Visual Query Tuning: Towards Effective Usage of Intermediate Representations for Parameter and Memory Efficient Transfer Learning
Decoupling Learning and Remembering: A Bilevel Memory Framework With Knowledge Projection for Task-Incremental Learning
Neuro-Modulated Hebbian Learning for Fully Test-Time Adaptation	➖
TIPI: Test Time Adaptation With Transformation Invariance			➖
Meta-Learning With a Geometry-Adaptive Preconditioner
Meta-Tuning Loss Functions and Data Augmentation for Few-Shot Object Detection	➖
A Probabilistic Framework for Lifelong Test-Time Adaptation
Few-Shot Class-Incremental Learning via Class-Aware Bilateral Distillation
CafeBoost: Causal Feature Boost To Eliminate Task-Induced Bias for Class Incremental Learning	➖		➖
A Strong Baseline for Generalized Few-Shot Semantic Segmentation			➖
Towards Better Stability and Adaptability: Improve Online Self-Training for Model Adaptation in Semantic Segmentation			➖
A New Benchmark: On the Utility of Synthetic Data With Blender for Bare Supervised Learning and Downstream Domain Adaptation
Cross-Image-Attention for Conditional Embeddings in Deep Metric Learning	➖
Principles of Forgetting in Domain-Incremental Semantic Segmentation in Adverse Weather Conditions			➖
Data-Free Knowledge Distillation via Feature Exchange and Activation Region Constraint
(ML)2P-Encoder: On Exploration of Channel-Class Correlation for Multi-Label Zero-Shot Learning
Finetune Like You Pretrain: Improved Finetuning of Zero-Shot Vision Models
Simulated Annealing in Early Layers Leads to Better Generalization
A Data-Based Perspective on Transfer Learning
Learning Expressive Prompting With Residuals for Vision Transformers	➖		➖
Boosting Transductive Few-Shot Fine-Tuning With Margin-Based Uncertainty Weighting and Probability Regularization	➖		➖
Improving Generalization With Domain Convex Game
Patch-Mix Transformer for Unsupervised Domain Adaptation: A Game Perspective
Guided Recommendation for Model Fine-Tuning	➖
Improving Generalization of Meta-Learning With Inverted Regularization at Inner-Level	➖
Hint-Aug: Drawing Hints From Foundation Vision Transformers Towards Boosted Few-Shot Parameter-Efficient Tuning